69th International Astronautical Congress, Bremen, Germany.

IAC–18–C4.6.2.x47764

Chemical Propulsion System Design for a 16U Interplanetary CubeSat

Karthik Venkatesh ManiPolitecnico di Milano, Italy

karthikvenkatesh.mani@polimi.it

Francesco TopputoPolitecnico di Milano, Italyfrancesco.topputo@polimi.it

Angelo CervoneDelft University of Technology, the Netherlands

a.cervone@tudelft.nl

Stand-alone interplanetary CubeSats to Mars require primary propulsion systems for orbit manoeuvring and tra-jectory control. The context is a 30 kg 16U Interplanetary CubeSat mission on a hybrid high-thrust & low-thrusttrajectory utilising a Dual Chemical-Electric Propulsion System to achieve Earth escape (high-thrust) and performautonomous deep-space cruise (low-thrust) to Mars. Mission analysis is performed to estimate the required ∆V(∼ 445 m/s) for Earth escape and Mars circularisation, and calculate the required thrusts and burn durations whileminimising gravity losses, Van Allen belt crossings, and irrecoverable destabilisation. Design of an ADN-based mono-propellant thruster and an Ethanol-H2O2 based bipropellant thruster are performed and the critical design detailsregarding propellant characteristics, system sizes, operating pressures, material selection and geometry of feed sys-tems, thrust chambers, and nozzles are delineated. Performances analysis of the monopropellant thruster yields amaximum thrust of 3.18 N (two thrusters with 1.54 N each) and an Isp = 259.7 s with nozzle area ratio of 100.Similarly, the bipropellant thruster yields a thrust of 2.91 N and an Isp = 303.15 s. The overall mass and volume ofthe monopropellant system are 5.6 kg (18.7%) and 7U, respectively. The bipropellant system weighs 4.85 kg (16.17%)while occupying 6.5U space.

1. Introduction

CubeSats have been used for Earth observationmissions since early 2000s with multiple applicationssuch as climate assessment, biological research, atmo-spheric characterisation etc. [1]. CubeSat design andmission development have been primarily pursued byuniversities and small-spacecraft consortia and thereis a growing interest among national and interna-tional space organisations. Contemporary CubeSatsoperate at Low-Earth Orbits (LEOs) and their sizerange is 1U to 6U while larger form factors are underdevelopment [2]. Earth-based CubeSats are designedto maximise payload performance and do not haveprimary propulsion systems for orbital manoeuvringsince such operations are not considered critical. Veryfew CubeSat missions have performed a technologydemonstration of propulsion systems, albeit with lim-ited capability. The IMPACT mission utilised multi-ple electrospray thrusters developed by MIT and theBricSAT-P mission utilised µCathode Arc Thrustersby George Washington University [3]. The propul-sion systems utilised in present-day CubeSats are for

attitude control, station-keeping, and reaction wheeldesaturation [4].

Interplanetary CubeSat missions expand the hori-zon of CubeSat applications and enable Solar Sys-tem exploration at a high science-to-investment ratio.Mission applications include Mars science observation[5], Mars communication relay and network setup[6, 7], asteroid mineral mapping, heliophysics stud-ies [8], lunar meteoroid observation [2] etc. Improve-ments in critical fields like long-distance communi-cation, deep-space autonomous guidance-navigation-control, propulsion for trajectory control, accurateADCS, and high-speed low-power on-board data pro-cessing are required [8].

CubeSat missions to Mars could be achievedthrough a) in-situ deployment by a mothership andb) highly flexible stand-alone CubeSats on deep-spacecruise [9]. The Mars Cube One (MarCO) mission,designed by JPL and launched alongside InSight lan-der mission in May 2018, is the only interplanetaryCubeSat in existence [6, 10]. The mission is injecteddirectly into the interplanetary space by the launcher.MarCO mission utilises two 6U CubeSats that shall

IAC–18–C4.6.2.x47764 Page 1 of 15

69th International Astronautical Congress, Bremen, Germany.

perform a Mars flyby and provide communication re-lay and support during InSight’s Mars atmosphericentry. The CubeSats use Vacco MiPS (Micro Cube-Sat Propulsion System) for executing two trajectorycontrol manoeuvres [11]. Stand-alone CubeSat mis-sions to Near-Earth Objects have been shown to befeasible, such as M-ARGO mission by ESA [12].

The stand-alone interplanetary CubeSat missionunder consideration in this work is the MarsAtmospheric Radiation Imaging Orbiter (MARIO)mission. It is envisaged to be launched into theSupersynchronous Geostationary Transfer Orbit (SS-GTO), execute orbit raising and Earth escape, per-form a low-thrust interplanetary transfer, achieveballistic capture at Mars, and then enter an areosyn-chronous orbit. Launch windows could be broadenedand the launch could be shared with any primarypayload that is bound for such a high-energy Earthorbit. Additionally, such launches have a higher fre-quency (∼10/year) compared to direct heliocentricorbit launches (∼2/year).

Robust primary propulsion systems become in-dispensable for interplanetary missions since precisetrajectory control and major orbital manoeuvre op-erations are required. Mani et al [9] have delin-eated the characteristics of the the dual chemical-electric propulsion systems design that is utilised inthe MARIO mission to execute a hybrid high-thrust& low-thrust trajectory. The motivation is to de-velop robust, compact, and powerful primary propul-sion systems that shall enable such interplanetarymissions and increase the mission flexibility and au-tonomy. This work focuses on the chemical propul-sion system which is used in the high-thrust stageto perform orbit raising manoeuvres and achieve aswift Earth escape to reduce the residence time inthe Van Allen radiation belts and consequently avoidradiation damage. The chemical propulsion systemis also utilised in performing a manoeuvre at Marsorbit to initiate the circularisation towards an are-osynchronous orbit.

Chemical propulsion systems applicable to in-terplanetary CubeSats include monopropellantthrusters [13, 14], bipropellant thrusters [15], tripro-pellant [16], cold gas [17, 3, 10], warm gas [18], solid[19] and hybrid motor systems [20]. These systemshave been manufactured and tested on-board largesatellites but they face multiple issues over miniatur-isation [21]. The monopropellant and bipropellantsystems are the most suitable for stand-alone inter-planetary CubeSat wanting to achieve Earth escape

due to their superior performance characteristics[17].

In this work, the mission characteristics are high-lighted in section 2. The chemical propulsion designstrategy is elucidated in section 3 and the designcharacteristics and performances of monopropellantand bipropellant systems are discussed in section 4and section 5, respectively. These sections highlightthe propellant properties, feed system design, thrustchamber & nozzle design and their corresponding per-formance values relating to thrust, specific impulse,velocities etc.

2. Mission Characteristics

The key phases of the MARIO mission are a) orbitraising & Earth escape, b) deep-space autonomouscruise, c) ballistic capture and d) areosychronous or-bit emplacement. The high-thrust chemical propul-sion system is utilised in the first phase to provide thenecessary ∆V for Earth escape. The spacecraft is in-serted into a highly elliptical super synchronous geo-stationary transfer orbit with a perigee of 295 km andan apogee of 90,000 km, which is used for some com-munication satellite launches. For instance, SpaceX’sFalcon 9 v1.1 rocket launched Thaicom 6∗ in January2014 in that orbit. Since the launch frequency of suchcommunication satellites is higher than Earth escapemissions, the selection of this initial orbit improvesthe launch opportunities. Additionally, owing to thehigh eccentricity of the orbit, the necessary escapevelocity at the perigee is significantly low. The orbitraising manoeuvres are illustrated in Fig. 1.

Fig. 1: Illustration of orbit raising manoeuvres andEarth escape

High thrust is used in this phase primarily due tothe presence of Van Allen radiation belts and the ex-pected multiple crossings of the spacecraft through

∗SpaceX Falcon 9 Data Sheet - Space Launch Reporthttp://www.spacelaunchreport.com/falcon9ft.html

IAC–18–C4.6.2.x47764 Page 2 of 15

69th International Astronautical Congress, Bremen, Germany.

these. Thus, it is essential to achieve a swift escapeand reduce the radiation damage. A low thrust op-tion will lead to excessive radiation damage due to asignificantly high residence time of the spacecraft inthe radiation belts.

The chemical propulsion system is necessary toprovide the required ∆V within a short duration.Owing to the size and mass of the spacecraft, 16Uand ∼30 kg, there is a limitation on maximum thrustsince very high thrusts, in tandem with thrust mis-alignment and engine misfire, might lead to irrecover-able destabilitsation. Additionally, lower thrust val-ues lead to longer thruster burn times, which resultsin higher gravity losses. Thus, the propulsion systemneeds to fire for a specific duration, in terms of burntime or burn arc about the perigee, and with a spe-cific thrust such that Earth escape is achieved swiftlywhile destabilisation, gravity losses, and radiation ac-cumulation are reduced.

The initial insertion orbit semi-major axis, eccen-tricity, inclination, right-ascension of the ascendingnode, argument of perigee, and true anomaly are [a,e, i, Ω, ω, θ] = [51526 km, 0.8705, 0.01◦, 0◦, 0◦, 0◦].The perigee altitude is 295 km and the apogee alti-tude is 90,000 km. The perigee velocity of the initiallaunch orbit is 10.57 km/s and the escape velocity atperigee point is 10.93 km/s. Thus, the ∆V for es-cape, at the perigee point considering an impulsivemanoeuvre, is ∼ 360 m/s.

2.1 Orbit raising manoeuvre strategies

Earth escape is achieved by executing multiple or-bit raising manoeuvres using the chemical propulsionsystem. Since real manoeuvres are considered, thethruster has to burn for a specific duration in eachorbit. A single burn manoeuvre to achieve escapewould drastically increase the gravity losses. Thus,the burns must be split prudently such that swift es-cape is achieved while gravity losses are controlled.The two strategies to execute the burn manoeuvresare a) constant burn time and b) constant burn arc.In the first strategy, a thrust (T ) & burn time (tb)combination that enables the fastest escape whileconsidering the system constraints is determined andin the second strategy, a thrust & burn arc (∆θb)combination is determined. These strategies help ussize and design the chemical propulsion system in ac-cordance with the mission operations.

The execution of the burn is split equally such thatit is initiated at tb/2 or ∆θb/2 before the perigee andit is completed at tb/2 or ∆θb/2 after the perigee,

as illustrated in Fig.2. This enables a more efficientthrottling manoeuvre and reduces gravity losses.

Fig. 2: Burn strategy with equal splitting of tb and∆θb about Perigee. The powered trajectory is redand the ballistic trajectory is blue.

To calculate the spacecraft trajectory for the con-stant burntime approach, a 2-body problem is as-sumed and the injection orbit parameters are ini-tialised in the beginning. The goal of the trajectoryanalysis is to find two things, a) the thrust-burntimecombination for the lowest flight time to escape andb) the required total ∆V and the corresponding pro-pellant mass spent in achieving that escape.

Different thrust-burntime combinations are sam-pled to find the T − tb combination that yields thelowest transfer time. Thus, a nested loop containing arange of thrusts and a range of burntimes is utilised.Within this nested loop, the procedure to calculatethe trajectory is implemented. In the constant burn-time approach, the Mean anomaly is calculated usingthe given burntime and the corresponding Eccentricanomaly is obtained. The true anomaly at which theburn is executed is calculated using Eq. (1)

θpb = −2 tan−1[√

1 + epb1− epb

tan

(Epb2

)](1)

The subscript ’pb’ indicates pre-burn values. Thetrue anomaly is denoted by θ, eccentricity is e, andthe Eccentric anomaly is E. The Keplerian orbitalelements before the burn are then converted to Carte-sian coordinates. The orbital equation of motion isintegrated over the burntime to obtain the burn tra-jectory.

r̈ +µ

r3r̂ =

~T

m(2)

ṁ = −

∣∣∣~T ∣∣∣Isp g0

(3)

~T = F · Tmax ·~v

||~v||(4)

IAC–18–C4.6.2.x47764 Page 3 of 15

69th International Astronautical Congress, Bremen, Germany.

Here r is the distance, µ is the gravitational parame-ter of Earth, m is the mass, ~T is the thrust vector, ~vis the instantaneous velocity vector, F is the throttle,and Tmax is the magnitude of maximum thrust. Oncethe burn trajectory is calculated, the orbital elementsat the end of the burn and for the next manoeuvre arecalculated. The ballistic flight time is calculated us-ing the eccentric anomalies Eab and Enm, where ’ab’and ’nm’ indicate after-burn and new-manoeuvre, re-spectively. The ballistic flight trajectory is calculatedby integrating the equations of motion over the bal-listic flight time, however while setting the thrust tozero.

The procedure to integrate the powered and ballis-tic trajectories is done until Earth escape is achieved.The total transfer time is calculated by summing upthe burn and ballistic durations of each orbit untilescape. Depending upon the thrust and burntime,the total transfer time and the number of manoeu-vres vary. In addition to this, the burn arc variesfor each burn since the orbital geometry varies butthe burntime is kept constant. The combination thatyields the lowest flight time to escape Earth and thelowest number of manoeuvres while adhering to thesystem constraints is selected.

In the constant burn arc strategy, the ∆θb is setfor each manoeuvre. The burn commences at−∆θb/2and ends at +∆θb/2 from the point of perigee. Thetime between them is calculated and the burn trajec-tory is obtained by integrating the equation of motionwith T as an input. A series of thrusts and burn arcsare sampled and the T −∆θb combination that yieldsthe lowest transfer time for Earth escape is selected.The burntime varies for each manoeuvre since ∆θb isheld constant.

For the constant burntime and constant burn arcstrategies, the overall propellant mass mp is calcu-lated by subtracting the final state mf from the ini-tial mass m0. The overall ∆V is calculated using theTsiolkovsky’s rocket equation (Eq. (5))

∆V = Isp g0 log

[m0mf

](5)

These parameters, along with the T − tb and T −∆θb values, govern the chemical propulsion systemdesign.

2.2 Circularisation burn at Mars

The chemical propulsion system is utilised not onlyfor Earth escape but also in the circularisation ma-noeuvre at Mars. Once the Earth escape is achieved,the chemical propulsion system is shut off and the

low-thrust electric propulsion system starts firing toenable the heliocentric transfer to Mars [9]. Thespacecraft executes a long duration electric propul-sion burn to reach Mars and enters a Martian orbitthrough Ballistic capture[22].

The capture is achieved solely by natural gravita-tional forces and it can also escape using the samemechanism. Additionally, the ballistic capture orbithas a very high eccentricity and it is unsuitable forscientific observation missions. Thus, circularisationto an areosynchronous orbit needs to be done. Per-forming the circularisation manoeuvre using only thelow-thrust electric propulsion leads to an extremelylong operational time. Thus, to expedite the circu-larisation, the chemical propulsion system is utilisedto provide a small ∆V (∼ 45 m/s) to reduce the ec-centricity and then the electric propulsion system isutilised to complete the operation.

3. System Design Strategy

3.1 Requirements

The primary requirements for the chemical propul-sion system for the MARIO mission that are estab-lished pertain to maximum thrust, ∆V , maximumburn duration, mass and safety. The propulsion sys-tem requirements are listed in Tab. 1

Tab. 1: Chemical Propulsion System Requirements

ID Type RequirementCPROP-01 PER The chemical propulsion

system shall provide a min-imum ∆V = 445 m/s fororbital transfer and circu-latisation manoeuvres.

CPROP-02 PER The chemical propulsionsystem shall have a maxi-mum thrust of 3 N

CPROP-03 PER The maximum thrustingtime of the chemical propul-sion system shall be 600seconds per orbital ma-noeuvre

CPROP-04 CON The total mass of the chem-ical propulsion system shallbe no more than 7.5 kg.

CPROP-05 CON The chemical propulsionsystem shall utilise non-toxic propellants

The rationale for CPROP-01 comes from the ∆V srequired for Earth escape and circularisation at Mars.

IAC–18–C4.6.2.x47764 Page 4 of 15

69th International Astronautical Congress, Bremen, Germany.

The initial orbit has a perigee altitude of 295 km andan apogee altitude of 90,000 km. The perigee ve-locity at the initial orbit is 10.5761 km/s and therequired escape velocity at that perigee altitude is10.9358 km/s. Thus, in the ideal case without losses,the ∆V required for escape is ∼360 m/s. A 10%margin is considered to include for gravity losses andother miscellaneous operational errors, thereby yield-ing a margined ∆Vesc of 396 m/s.

To expedite the circularisation operation at Marsto an areostationary orbit from an initial high eccen-tricity ballistic capture orbit, a deceleration manoeu-vre at the periareion is executed. The decelerationmanoeuvre requires a ∆V of 45 m/s [23]. An addi-tional ∼10% margin is placed and the ∆Vcirc is cal-culated to be 49 m/s. Thus, the overall minimum∆Vtot that is required to be provided by the chemicalpropulsion system is set at 396 + 49 = 445 m/s.

Requirement CPROP-02 establishes a limitationon thrust and CPROP-03 establishes the maximumburn time. For a 30 kg spacecraft, the combinationof maximum thrust and burn durations are set to ef-fectively distribute the ∆V required to escape Earthinto multiple manoeuvres and to reduce the overalltransfer time while keeping the gravity loss in control.In contrast, a single burn manoeuvre to escape leadsto a gravity loss of ∼23%. Additionally, the limita-tions on thrust and burntime lead to the avoidance ofirrecoverable destabilisation and excessive heat buildup. Considering a 1 mm misalignment, the torqueis 3 mNm and the momentum is 1.8 Nms. Simulta-neous thrust vectoring and reaction wheel operationare essential for controlling the momentum build-upand proper directional firing. The maximum massrequirement, CPROP-04, is capped at 25% of theoverall spacecraft mass. The CubeSat is assumedto be a secondary payload and non-toxic propellantsneed to be used to avoid any damage to the primaryspacecraft as well as any self-damage. The toxic pro-pellants require rigorous containment strategies thatincrease the system cost. This requirement can berelaxed if the propellant has a low-toxicity and af-fordable control measures can be implemented.

3.2 System types

The types of chemical propulsion system underconsideration are solid, liquid, and cold gas. The keydesign driver is the specific impulse, Isp, which is thehighest for liquid thrusters and lowest for cold gasthrusters. In addition to this, the cold gas thrustersfor cubesats do not satisfy requirements on minimumthrust (< 1 N), which lead to very long flight times,

and maximum mass (> 8 kg). The solid thrustershave a lower Isp than liquid thrusters but are able toprovide a higher thrust. However, since multiple ma-noeuvres are required, start-stop capability is indis-pensable for the chemical propulsion system. Thus,the choice of the propulsion type is liquid propulsion.

In liquid chemical propulsion systems, the specificimpulse could vary between 200 s to 350 s, depend-ing upon the type of the propellants. Two propul-sion system types are considered: Monopropellantthrusters and Bipropellant thrusters. The mono-propellant thrusters that use ammonium dinitramide(ADN) blends are considered since they satisfy thenon-toxicity requirement (CPROP-05) and have highdensity. The specific impulse of the monopropellantsystem is 260 seconds and an ADN-based monopro-pellant system has been successfully demonstrated inspace[24, 25]. The bipropellant thrusters that utilisehydrogen peroxide (H2O2) and ethanol (C2H5OH)also satisfy CPROP-05. The specific impulse of suchbipropellant thrusters is 315 s [15].

3.3 Design procedure

The design procedure involves the selection ofthrust and burn durations for both the monopropel-lant and bipropellant thrusters. This is based on aparametric analysis to calculate the trajectory, thetotal time for Earth escape through orbit raising, andthe required propellant mass for the operations (seesection 2.1). Depending on the Isp, the total flighttime, propellant mass and consequently the thrustand burn durations vary†. The propulsion systemsizing is done using the mission requirements, the sys-tem requirements and system type. Propellant char-acteristics are analysed and the feed system is de-signed. The thrust chamber and nozzle design takeinto account the combustion properties and perfor-mance requirements.

4. Monopropellant thruster

The parametric analysis to calculate the thrustand burn duration values based on minimum transfertime to escape Earth is performed. The monopropel-lant Isp is set at 260 seconds while the minimum andmaximum thrust values are set at 2N and 3N, re-spectively. For the constant burntime strategy, thesampled values of tb range from 400 to 600 seconds.For the constant burnarc strategy, the sampled val-ues of ∆θb range from 35

◦ to 55◦ of total arc. To find

†All thrust and specific impulse values used in this workpertain to vacuum condition

IAC–18–C4.6.2.x47764 Page 5 of 15

69th International Astronautical Congress, Bremen, Germany.

the precise thrust and burn duration, 1000 values ofthrust, tb and ∆θb are sampled.

The total transfer time or the total Earth orbitingtime as a function of T and tb is illustrated in Fig. 3afor the constant burntime strategy. The T−tb combi-nation is picked from the zone where the transfer timeis the lowest (top-right, dark blue). Some combina-tions yield a transfer time that is greater than 150days (3600 hours) and they are discarded because aswift transfer is desired. Similarly, Fig. 3b illustratesthe total transfer time as a function of T and ∆θb(burn arc).

(a) Transfer time (hr) for constant tb strategy

(b) Transfer time (hr) for constant ∆θb strategy

Fig. 3: Transfer time (hr) as a function of T & tband T & ∆θb for Isp = 260 s

If the burntime is kept constant for each manoeu-vre, the thrust T = 2.87 N, burntime tb = 589.5792 s,and the lowest total transfer time P = 764.65 hours.The maximum and minimum burn arcs correspond-ing to this tb are 53.19

◦ and 52◦. If the burn arc isset constant for each manoeuvre, T = 2.77 N, ∆θb =53.67◦ and the lowest transfer time P = 757.92 hours.The maximum burntime corresponding to this ∆θb is611.3 seconds, which violates CPROP-03, albeit by asmall amount.

These parameters indicate the thrust that needs tobe provided and the burn durations for which eachmanoeuvre should be executed to achieve the low-

est transfer time. However, the design of thrustercannot be done precisely for this thrust and somemargins and contingency measures need to be taken.Thus, the thruster is designed for a thrust of 3 Nand a robust control authority can then be providedto the thruster to execute such precise manoeuvres.Thruster firing at 3 N reduces the tb for each ma-noeuvre by ∼4.5% and the transfer time increases by< 1%. Similarly, the ∆θb for each manoeuvre re-duces by 7.3% and the corresponding transfer timeincreases by < 1%.

The constant burntime strategy using monopro-pellant thrusters firing at T = 2.87 N for tb = 589.57seconds yields a total ∆V = 363.041 m/s and a pro-pellant mass mp = 3.96 kg. The ideal ∆V for escapeat perigee while considering impulsive manoeuvres is359.66 m/s. Thus, the difference between the ∆V sof real and ideal manoeuvres is 3.38 m/s, which is∼1% higher than the ideal one. The constant burnarc strategy using the thrusters firing at T = 2.77N for a ∆θb = 53.67

◦ yields a total ∆V = 363.29m/s and a propellant mass mp = 3.96 kg. This extra∆V accounts for the gravity losses accumulated overmultiple manoeuvres and it is well within the appliedmargin of 10%. The propellant mass (mp,mg) pertain-ing to the margined value (∆Vmg = 396 m/s) is 4.32kg. Another 0.38 kg of propellant mass is added to ac-commodate the Mars circularisation ∆Vcirc,mg = 49m/s. Both strategies lead to an escape after the 6th

orbit, bringing the total number of Van Allen beltcrossings to 13 (12 elliptical crossings plus 1 paraboliccrossing). The orbital transfer is illustrated in Fig. 4

Fig. 4: Orbit raising using monopropellant thrusterswith constant tb strategy

To provide the necessary thrust, either a single 3Nthruster or two 1.5N thrusters can be utilised. ADN-based thrusters providing 1 - 1.5N thrust are cur-rently under development [25, 26]. Thus, this analysisconsiders two 1.5N thrusters to provide the necessarymaximum thrust. The mission parameters are listedin Tab. 2

IAC–18–C4.6.2.x47764 Page 6 of 15

69th International Astronautical Congress, Bremen, Germany.

Tab. 2: Orbit raising parameters using monopropel-lant thruster with constant tb strategy

Parameter Value

Initial Mass, mi 30 kgRequired Thrust, T 2.87 NSpecific Impulse, Isp 260 sBurn Time per manoeuvre, tb 589.58 sOrbit raising manoeuvres 6Ideal ∆V 359.66 m/sReal ∆V 363.041 m/sMargined ∆V (10% margin) 396 m/sPropellant mass (real), mp 3.96 kgPropellant mass (margined), mp,mg 4.32 kgFinal mass at Earth escape, mf,esc 26.04 kgMars circularisation ∆Vcirc 45 m/sMargined ∆Vcirc,mg (∼10% margin) 49 m/sPropellant mass for ∆Vcirc,mg 0.38 kgFinal mass at Mars, mf,circ 19.31 kg

1

Overall propellant mass, mp,mg,ovr 4.7 kg1 After low-thrust transfer to Mars and chemical circularisa-

tion manoeuvre

4.1 Propellant properties

The propellant considered is an ammonium di-natramide (ADN),NH4[N(NO2)2], blend propellant -FLP-106 with a liquid phase density of 1357 kg/m3

at 25◦C. ADN is a solid white salt with ammoniacation (NH4+) and dinitramide anion (N(NO2)2−)which is readily soluble in water and other polar sol-vents. FLP-106 consists of 64.6% ADN, 23.9 % waterand 11.5% low-volatile fuel [27]. In comparison, thewidely used monopropellant Hydrazine has a densityof 1.004 kg/m3 at room temperature and has a veryhigh toxicity. At nozzle expansion ratio of 50, theFLP-106 propellant can yield an Isp ≈ 260 s [28].Some key properties of FLP-106 are listed in Tab. 3[27, 28].

Tab. 3: FLP-106 liquid monopropellant properties

Property Value

Molecular mass, M 22.8 kg/kmolDensity, ρ 1357 kg/m3

Heat capacity, cp 2.41 kJ/kg ·KThermal expansion coeff.,α 6.04× 10−4 1/KSaturation temperature, Ts 273.15 KDynamic viscosity, µ 3.7 mPa · sVapour pressure, pv < 21 mbar

The propellant is produced by dissolving the solidADN salt in fuel/water mixture. The dissolution ofADN in the mixture results in a drop in tempera-ture and a slowing down of the process, which canthen be speeded up by treating it to a warm waterbath [28]. Eventhough many propellants are classifiedwithin hazard 1 class explosive materials and have re-strictions on transportation, the propellant FLP-106should not fall into this category due to its very lowsensitivity and volatility [27].

The method of ignition and the ignition charac-teristics of the propellant play an important role inpropulsion system design. ADN-blend propellants,which contain water, have a significant ignition delaydue to water evaporation. Additionally, a hot surfaceis required to sustain the propellant decomposition.Wilhelm et al have found that thermal ignition usinga glowplug yields a stable flame and an acceptableignition time for FLP-106 propellant [29]. The glow-plug needs to be preheated until it reaches a temper-ature of 900◦C. The combustion temperatures rangefrom 1600◦C to 1900◦C depending upon the cham-ber pressure [28, 29]. The temperature at which theADN decomposition starts is 150◦C [30]. The ADNundergoes a complete thermal decomposition beforeentering the main combustion chamber.

4.2 Feed system

The feed system consists of the storage tanks,valves, flow lines, and the tank pressurization sys-tem. The margined propellant mass, mp,mg = 4.7 kgfor the monopropellant system with an Isp = 260 s.Considering the liquid phase density of FLP-106 atroom temperature (see Tab. 3), the propellant vol-ume Vp = 3464 cm

3, which in terms of CubeSat unitsis ∼ 3.5 U.

Since the maximum thrust requirement is low com-pared to the traditional satellite propulsion systems,a pressure-fed system is utilised. However, since thethrust must be precise and constant, a regulated pres-sure feed is utilised rather than the classical blow-down system. FLP-106 has low volatility and lowsensitivity, thus GN2 is used as the pressurant. TheGN2 is stored in a small tank and supplied to themain propellant tanks through a pressure regulatorand a check valve.

A pressurised propellant tank requires a spheri-cal or a cylindrical shape with hemispherical ends forsafety reasons. However, for a 3464 cm3 tank, the ra-dius of the spherical tank will be approximately 9.45cm, which cannot be accommodated easily in a 16UCubeSat. To minimise the tank volume while main-

IAC–18–C4.6.2.x47764 Page 7 of 15

69th International Astronautical Congress, Bremen, Germany.

taining the pressure and accommodating within thestructure, cylindrical tanks with ends containing el-liptical dome shapes have to be developed. Since onelarge tank would be infeasible, four tanks with 980cm3 capacity each, including a ∼ 13% ullage volume,are utilised. Each tank has a radius of 4.7 cm and aheight of 16.4 cm. The tanks shape and the place-ment inside the spacecraft structure are illustrated inFig. 5. Each tank occupies ∼ 1.6U space.

(a) Tank shape (b) Full structure

Fig. 5: Tank shape and placement in the spacecraftstructure

The feed pressure is set at 2.2 MPa and this yieldsa Maximum Expected Operational Pressure (MEOP)of 2.6 MPa [31]. The feed pressure value is similarto the one for a high-performance ADN-based mono-propulsion system on-board PRISMA satellite. Thethickness of the tank wall is determined for a burstpressure of 3.9 MPa, which considers a burst factorof 1.5 applied on MEOP.

The material used for the tank is Titanium alloyTi-6Al-4V ‡ which has an yield strength of 880 MPaand an ultimate strength of 950 MPa. It is widelyused as an implant material and has a very high cor-rosion resistance. The density of the material is 4430kg/m3. The propellant FLP-106 is also compatiblewith this material [27]. The tank wall thickness iscalculated to be 0.2 mm. Laser-assisted machiningand other precision techniques applied on Ti-6Al-4Venable the manufacturing of tanks with small thick-nesses [32, 33]. The total mass of all tanks is 0.172 kg.If the a steel alloy material AISI4140 with a densityof 7850 kg/m3 and yield strength of 415 MPa is used,the tank thickness is 0.45 mm and the overall tankmass is 0.68 kg, which is 4 times that of the Ti-alloy.

The pressurant gas (GN2) pressure is consideredto be 28 MPa at 323 K [34]. The nitrogen gas den-sity at these conditions is 257.6 kg/m3. The pressur-ant gas volume and mass are calculated based on thefeed pressure, 2.2 MPa, and feed temperature, 300

‡http://asm.matweb.com/search/SpecificMaterial.asp?bassnum=mtp641

K. The GN2 tank volume is 415.6 cm3 and the tank

dimensions are rgas,tank = 4.7 cm and hgas,tank = 6cm. Utilising the same Titanium alloy material, thegaseous tank thickness is 2.24 mm and the tank massis 0.18 kg. The gas tank occupies ∼ 0.6U space in thespacecraft. The feed system design characteristics arelisted in Tab. 4. The feed system mass does not in-clude the propellant mass but includes the pressurantgas mass.

Tab. 4: Monopropellant thruster feed system designparameters

Property Value

Propellant tanks 4Individual tank volume, Vtank 980 cm

3

Prop. feed pressure, Pprop 2.2 MPaProp. tank burst pressure, Pprop,burst 3.9 MPaProp. tank radius, rprop,tank 4.7 cmProp. tank height, hprop,tank 16.4 cmProp. tank thickness, thprop 0.2 mmTotal tank mass, mtank 0.172 kgPress. gas pressure, Pgas 28 MPaPress. gas mass, mgas 0.107 kgPress. tank volume, Vgas,tank 415.6 cm

3

Press. tank radius, rgas,tank 4.7 cmPress. tank height, hgas,tank 6 cmPress. tank thickness, thgas,tank 2.24 mmPress. tank mass, mgas,tank 0.18 kgFeed pipes & valves mass, mfv 0.2 kgFeed system total mass, mfeed 0.66 kgFeed system total volume 7U

4.3 Thruster design and performance

Some studies utilise a hot catalytic bed for the de-composition of ADN [35]. FLP-106 ignites at 200◦Ctemperature [27, 29]. The catalyst bed design is acomplicated process and it has not been addressedin this work. In addition to catalystic decomposi-tion and ignition of ADN-based propellants, thermalignition using a glow-plug and ignition using electri-cal resistive heating are also functional [29, 30]. Inthermal ignition technique, the glow-plug is suppliedwith power for preheating and sustenance of the hotsurface at 900◦C. The FLP-106 propellant makes con-tact with the surface and ignition is initiated. Sincea complete ADN decomposition is assumed, the ther-mal decomposition reaction efficiency ηr = 1.

The chamber pressure is 2 MPa and the combus-tion temperature is 1880 K [28, 24]. The specific heatratio of the combustion products k = 1.15 and the

IAC–18–C4.6.2.x47764 Page 8 of 15

69th International Astronautical Congress, Bremen, Germany.

specific gas constant Rgas = 364.7 J/kg ·K, whichis calculated using the Molecular Mass M = 22.8kg/kmol. The nozzle throat diameter, Dt, is set at0.5 mm and the effect of the nozzle area ratio on theperformance is analysed by varying it from ε = 50 toε = 100. The nozzle exit pressure, Pe, is determinedusing the Eq. 6 [36].

AtAe

=

(k + 1

2

)1/(k−1)·(PePc

)1/k·√√√√(k + 1

k − 1

)·

(1−

(PePc

)(k−1)/k)(6)

Here, Ae and At are the nozzle exit and throatareas, respectively. The nozzle is assumed to be a su-personic convergent-divergent conical nozzle and theMach number at the throat is 1. The mass flow rateṁ, the coefficient of thrust CF , and the characteristicvelocity c∗ are calculated using Eqs. 7,8 and 9, whichare ideal rocket relations. Since the thruster opera-tion occurs in near vacuum conditions, Pvac is set as0.

ṁ =At · Pc · k√

(k ·Rgas · Tc)·

√(2

k + 1

) k+1k−1

(7)

CF,ideal =

√√√√2 · k2k − 1

·(

2

k + 1

) k+1k−1

·

(1−

(PePc

) k−1k

)

+

(Pe − Pvac

Pc

)·(AeAt

)(8)

c∗ideal =1

k·

√k ·Rgas · Tc ·

(k + 1

2

) k+1k−1

(9)

The nozzle efficiency ηn is assumed to be 0.97 and thereaction efficiency is set at 1 since there is completedecomposition. Thus, the corrected values become,

CF,corr = CF,ideal · ηn (10)c∗corr = c

∗ideal · ηr (11)

Utilising Eqs.7-11, the thrust and specific impulseare calculated using Eqs.12 and 13.

T = ṁ · c∗corr · CF,corr (12)

Isp = CF,corr ·c∗corrg0

(13)

Considering one thruster, for ε = 50, Pc = 2 MPa,Tc = 1880 K, and Dt = 0.5 mm, the performancevalues are ṁ = 0.606 g/s, CF,ideal = 1.964, c

∗ideal =

1296 m/s, Tideal = 1.542 N, and Isp,ideal = 259.5 s.With ηn = 0.97, the corrected and estimated perfor-mance values are CF = 1.905, T = 1.496 N and theIsp = 251.7 s. Since two thrusters are used, systemis capable of providing Tmax,ideal = 3.18 N, with anestimated Tmax = 2.992 N considering the nozzle ef-ficiency. The thrust and the specific impulse increasewith the increase in area ratio, as illustrated in Fig. 6.At ε = 100, Tideal = 1.591 N and Isp,ideal = 267.7 s,and the T = 1.543 N and Isp = 259.7 s. The finalvalue of ε = 100 is chosen for the design.

Fig. 6: Isp and T variation with Area Ratio for themonopropellant thruster with ηn = 0.97

The thruster design parameters such as chambervolume Vc and chamber length Lc are calculated us-ing the desired residence time ts (see Eqs. 14- 15).The desired residence time ts is set at 0.001 s to ac-count for ignition and decomposition. The contrac-tion ratio of the nozzle is set at 50 and the chamberdiameter Dc is calculated to be 10 ×Dt. The nozzleconstriction length Lcon is set as 1.5 mm as a prelim-inary design choice so as to adjust the constrictionhalf angle to 45◦.

Vc = ts ·At√k ·Rgas · Tc ·

(2

k + 1

)1/(k−1)(14)

Vc = Ac · Lc +Ac · Lcon ·

(1 +

√AtAc

+AtAc

)(15)

For the set input parameters, the Vc = 215.3 mm3

with Lc = 9.223 mm. The Ac and At are chamberand throat cross section areas, respectively.

IAC–18–C4.6.2.x47764 Page 9 of 15

69th International Astronautical Congress, Bremen, Germany.

A conical nozzle with a half angle α = 15◦, ε = 100,and the throat longitudinal radius is set at 1.5×Rt.The length of the divergent part of the nozzle is deter-mined to be LN = 8.45 mm. The chamber and nozzledesign parameters are listed in Tab. 5 and the overallperformance and system design parameters are listedin Tab. 6. The thrusters are placed outside the 16Uspacecraft structure.

Tab. 5: Monopropellant thruster chamber & nozzledesign parameters

Property Value

Thrusters 2Chamber pressure, Pc 2 MPaChamber temperature, Tc 1880 KChamber volume, Vc 215.3 mm

3

Chamber diameter, Dc 5 mmChamber residence time, ts 0.001 sChamber length, Lc 9.223 mmExpansion area ratio, ε 100Expansion half angle, α 15◦

Nozzle length (divergent), LN 8.45 mmIgniter mass, mig 0.02 kgThruster mass, mthruster 0.2 kg

Tab. 6: Monopropellant thruster overall perfor-mance and design parameters

Property Value

Decomposition efficiency, ηr 1Nozzle efficiency, ηn 0.97Max Thrust (per thruster) 1.543 NSpecific Impulse, Isp 259.7 sExhaust velocity, ve 2548 m/sCharacteristic velocity, c∗ 1296 m/sThrust coefficient, CF 1.905Mass flow rate (per thruster), ṁ 0.606 g/sMax temperature, Tmax 1880 KOverall Mass, mtot 5.6 kg

5. Bipropellant thruster

The parametric analysis to find the thrust andburn duration values based on minimum transfer timeto escape Earth is performed again for the bipropel-lant thruster system with an Isp = 315 s. The sam-pled thrusts are between 2 N and 3 N. For the con-stant burntime strategy, the values of tb range from400 to 600 seconds while for the constant burn arc

strategy, the values of ∆θb range from 35◦ to 55◦. For

each combination, the trajectory is simulated and thetotal transfer time is predicted.

The transfer time as a function of T and tb is il-lustrated in Fig. 7 and similarly, Fig. illustrates thetransfer time as a function of T and ∆θb. The com-binations are selected based on the lowest transfertime to escape. The combinations that yield a totaltransfer time greater than 3600 hours are discarded.

(a) Transfer time (hr) for constant tb strategy

(b) Transfer time (hr) for constant ∆θb strategy

Fig. 7: Transfer time (hr) as a function of T & tband T & ∆θb for Isp = 315 s

The constant burntime strategy yields a combina-tion of T = 2.882 N and tb = 594.39 s for each ma-noeuvre. The shortest total transfer time P = 774.47hours. The maximum and minimum burn arcs dur-ing the transfer are 53.6◦ and 52.40◦, respectively.The constant burn arc strategy yields a combinationof T = 2.738 N and ∆θb = 54.9

◦. The total transfertime P = 767.36 hours. The corresponding maximumand minimum burntimes are 626.24 s and 390.67 s.Manoeuvring at a maximum thrust of 3 N, the tbbecomes 570.51 s, which is a ∼4% decrease and thetransfer time increases by < 1%. For the constantburn arc strategy, the ∆θb becomes 50.3

◦, which isan 8.3% decrease. Similar to the monopropellantthruster, the bipropellant thruster is also sized forthe maximum thrust.

IAC–18–C4.6.2.x47764 Page 10 of 15

69th International Astronautical Congress, Bremen, Germany.

For a propulsion system with Isp = 315, the total∆V for Earth escape is 361.31 m/s and the corre-sponding propellant mass, mp = 3.31 kg. The space-craft escapes after 6 manoeuvres and this ∆V is accu-mulated over these manoeuvres. The ∆V varies fromthe ideal perigee escape ∆V , 359.66 m/s, by 1% andaccounts for the gravity losses. The propulsion sys-tem is sized for a margined ∆Vmg of 445 m/s, whichincludes the margined velocity increment to escape,∆Vmg = 396 m/s and the margined circularisation∆Vcirc = 49 m/S. The propellant mass, mp,mg, cor-responding to ∆Vmg is 3.61 kg and the propellantmass for circularisation is 0.30 kg. Thus, the over-all propellant mass mp,mg,ovr = 3.91 kg. A singlethruster providing a 3 N thrust is considered. Theorbit raising parameters using bipropellant thrusterwith constant tb strategy are listed in Tab. 7.

Tab. 7: Orbit raising parameters using bipropellantthruster with constant tb strategy

Parameter Value

Initial Mass, mi 30 kgRequired Thrust, T 2.882 NSpecific Impulse, Isp 315 sBurn Time per manoeuvre, tb 594.39 sOrbit raising manoeuvres 6Ideal ∆V 359.66 m/sReal ∆V 361.31 m/sMargined ∆V (10% margin) 396 m/sPropellant mass (real), mp 3.31 kgPropellant mass (margined), mp,mg 3.61 kgFinal mass at Earth escape, mf,esc 26.39 kgMargined ∆Vcirc,mg (∼10% margin) 49 m/sPropellant mass for ∆Vcirc,mg 0.30 kgOverall propellant mass, mp,mg,ovr 3.91 kg

5.1 Propellant properties

Bipropellant thrusters utilising ethanol and hydro-gen peroxide for small spacecraft applications havebeen under development recently [15, 37]. The fuel,ethanol, has a density of 789.3 kg/m3, molar mass of49.07 kg/kmol, and a vapour pressure of 5.95 kPa at293 K. Ethanol freezes at 159 K and vapourises at351 K.

The oxidiser is a concentrated hydrogen perox-ide solution or ”high-test peroxide” (HTP) which is95% w/w H2O2 in water. Although H2O2 solution isnon-flammable and stable when stored properly[38],it readily decomposes into a high temperature (∼700◦C) mixture of steam and oxygen in contact with

a catalyst. The 95% H2O2 solution has a densityof 1420 kg/m3 at 293 K. The solution has a molarmass of 33.21 kg/kmol with pure H2O2 having 34.01kg/kmol. The melting point is 272.7 K and the boil-ing point is 423.35 K.

Both propellants satisfy the requirement CPROP-05 regarding non-toxic propellant usage. Howevercare must be taken while handling highly concen-trated H2O2 since a skin contact might lead to ex-tensive burns. Additionally, only a few materials arecompatible with HTP for long term storage [38]. Themetallic or alloy based storage tank material almostalways require passivation and conditioning beforeuse. The Titanium alloy (Ti-6Al-4V) and stainlesssteel coated with Polytetrafluoroethylene (PTFE) arecompatible with HTP [38].

5.2 Feed System

The bipropellant feed system contains separatestorage tanks for ethanol and H2O2. The systemalso includes a GN2 tank for propellant pressurisa-tion and separate flow lines and flow control valves.The margined propellant mass mp,mg = 3.91 kg forIsp = 315 s, which includes the propellant masses forproviding the required ∆V for Earth escape and Marscircularisation.

The feed pressure is set at 1.8 MPa and the com-bustion chamber pressure is assumed to be 1.5 MPa[39]. The mass mixture ratio (O/F) is fixed at 4.5 toensure a fuel-rich condition during combustion sincethe vacuum specific impulse Isp,vac is near the maxi-mum value. Eventhough O/F = 4.4 yields the high-est Isp, O/F 4.5 is chosen to reduce the heat releasewhile obtaining a similar performance. Fig. 8 repre-sents the Isp and T variation with the O/F ratio andFig. 9 represents the combustion temperature withthe O/F ratio.

The fuel mass, mfuel = 0.711 kg and the oxidisermass mox 3.199 kg. Considering the densities of thepropellants at 293 K, the volumes of fuel and oxidiserare Vfuel = 901 cm

3 and Vox = 2253 cm3, respectively.

Following the same principles to accommodate thepropellant tanks in the structure, the fuel is stored inone 1018 cm3 dome shaped cylindrical tank and theoxidiser is stored in three 848.6 cm3 tanks, each with13% ullage volume. Each tank has a radius of 4.7cm. The height of the fuel tank hfuel,tank = 17 cm,occupying 1.7U space. The height of each oxidisertank hox,tank = 14.3 cm and it occupies ∼ 1.4U.

With the feed pressure at 1.8 MPa and the GN2pressurant is set at 28 MPa at 323 K [34]. The re-quired GN2 mass to pressurise all tanks is 0.085 kg.

IAC–18–C4.6.2.x47764 Page 11 of 15

69th International Astronautical Congress, Bremen, Germany.

Fig. 8: Theoretical Isp and T variation with O/F;Pc = 1.5 MPa, ε = 100, Dt = 1.15 mm

Fig. 9: Combustion Temperature variation withO/F; Pc = 1.5 MPa

The GN2 tank has a volume of 329.5 cm3, with a

radius of 4.7 cm and a height of 5.45 cm, occupying∼0.5U space. The tank material used is Ti-6Al-4V al-loy. The tank MEOP is 2.16 MPa and the tank burstpressure is 3.24 MPa. The thickness of each tank is0.173 mm. The fuel tank mass mfuel,tank = 0.0333kg and the combined oxidiser tank mass mox,tank =0.0832 kg. The GN2 dome shaped tank thicknessis 2.24 mm and the corresponding mass mgas,tank =0.143 kg. The feed system characteristics are listedin Tab. 8.

5.3 Thruster design and performance

The high-test peroxide (HTP) or the 95% H2O2 so-lution undergoes an exothermic decomposition whenin contact with a catalyst. The resulting vapour read-ily ignites in contact with a fuel depending upon theconditions in the combustion chamber. The catalystis placed in a small catalytic chamber outside the

Tab. 8: Bipropellant thruster feed system design pa-rameters

Property Value

Fuel tanks 1Oxidiser tanks 3Fuel tank volume, Vfuel,tank 1018 cm

3

Ox. tank volume, Vox,tank 848.6 cm3

Feed pressure, Pprop 1.8 MPaTank burst pressure, Pprop,burst 3.24 MPaTank radius, rprop,tank 4.7 cmFuel tank height, hfuel,tank 17 cmOx. tank height, hox,tank 14.3 cmTank thickness, thprop 0.173 mmFuel tank mass, mfuel,tank 0.0333 kgOx. tank mass (total), mox,tank 0.0832 kgPress. gas pressure, Pgas 28 MPaPress. gas mass, mgas 0.163 kgPress. tank volume, Vgas,tank 329.5 cm

3

Press. tank radius, rgas,tank 4.7 cmPress. tank height, hgas,tank 5.45 cmPress. tank thickness, thgas,tank 2.24 mmPress. tank mass, mgas,tank 0.143 kgFeed pipes & valves mass, mfv 0.2 kgFeed system total mass, mfeed 0.622 kgFeed system total volume 6.5 U

main thrust chamber and HTP is fed through. Thechosen catalyst is composed of a Cordierite substratewith Platinum Oxide washcoating, where Pt remainsas active phase [37]. Although auto-ignition takesplace, an electrical spark igniter is placed to ensureproper ignition.

At Pc = 1.5 MPa, the combustion of ethanol and95% H2O2 yields a combustion temperature Tc =2673.9 K. The specific heat ratio k = 1.139 and thecombustion gas constant Rgas = 381 J/kg ·K. Thenozzle area ratio is set at ε = 100 and the throat di-ameter is set at Dt = 1.15 mm. The resulting massflow rate ṁ = 0.9792 g/s. The ideal performance pa-rameters are CF,ideal = 1.999, c

∗ideal = 1590.96 m/s,

ve,ideal = 3181.78 m/s, Tideal = 3.114 N and Isp,ideal= 324.45 s. Rocket Propulsion Analysis tool devel-oped by Ponomarenko et al has been utilised to calcu-late the reaction and nozzle efficiencies as well as theestimated delivered performance of the thruster [40].The Ethanol-HTP (both at 293 K) reaction efficiencyηr = 0.9612, the nozzle efficiency ηn = 0.9721, andthe total efficiency ηt = 0.9344. The corrected per-formance values are, CF = 1.944, c

∗ = 1529.25 m/s,ve = 2972.92 m/s, T = 2.911 N and Isp = 303.15 s.

IAC–18–C4.6.2.x47764 Page 12 of 15

69th International Astronautical Congress, Bremen, Germany.

Considering the estimated performance, the propel-lant mass required for escape increases by 0.12 kg.The variation of T and Isp with the area ratio for thebipropellant thruster is illustrated in Fig. 10.

Fig. 10: Estimated Isp and T variation with AreaRatio for the bipropellant thruster; At = 1.039mm2, Pc = 1.5 MPa, ηtot = 0.9344

The nozzle contraction ratio is set at 50 and thechamber diameter Dc is calculated to be 8.132 mm,which is ∼7 times larger than Dt. A characteristiclength (L∗) of 1.78 m is chosen for the preliminarythruster design [41, 42]. The L∗ = 1.78 m corre-sponds to H2O2-RP1 combination including the cat-alyst bed and has been utilised in this case since thevalue for H2O2-Ethanol could not be found. The cor-responding residence time ts = 0.00268 s. The nozzleconstriction length is set at Lc = 1.6 mm. The re-sulting combustion chamber volume Vc = 1849 mm

3

and the combustion chamber length Lc = 33.74 mm.With an expansion half angle α = 15◦ and the throatlongitudinal radius is set at 0.75×Dt [41]. The lengthof the divergent part of the nozzle LN = 19.43 mm forε = 100. The chamber and nozzle design parametersare listed in Tab. 9 and the performance parametersare listed in Tab. 10.

The thrust chamber material selection is criticallyimportant since the combustion temperature is veryhigh. Niobium alloy is a good candidate since it hasa melting temperature of 2741 K and a density of8570 kg/m3. The melting temperature is marginallyhigher than the combustion temperature 2673.9 K.However, the safe operating temperature of the ma-terial is 1500 K and a fuel film cooling technique mustbe utilised to prevent overheating and structural fail-ure [39, 43]. Injecting a coolant at the location of themain fuel injector ensures high cooling capacity andsimplicity in design [39].

Tab. 9: Bipropellant thruster chamber & nozzle de-sign parameters

Property Value

Chamber pressure, Pc 1.5 MPaChamber temperature, Tc 2673.9 KChamber volume, Vc 1849 mm

3

Chamber diameter, Dc 8.132 mmCharacteristic length, L∗ 1.78 mChamber residence time, ts 0.00268 sChamber length, Lc 33.74 mmExpansion area ratio, ε 100Expansion half angle, α 15◦

Nozzle length (divergent), LN 19.42 mmCatalytic chamber mass, mcat 0.02 kgThruster mass, mthruster 0.3 kg

Tab. 10: Bipropellant thruster overall performanceand design parameters

Property Value

Reaction efficiency, ηr 0.9612Nozzle efficiency, ηn 0.9721Max Thrust (per thruster) 2.911 NSpecific Impulse, Isp 303.15 sExhaust velocity, ve 2972.92 m/sCharacteristic velocity, c∗ 1529.25 m/sThrust coefficient, CF 1.9441Mass flow rate, ṁ 0.9792 g/sMax temperature, Tmax 2673.9 KOverall Mass, mtot 4.852 kg

6. Conclusion

This work set out to design and characterise chemi-cal propulsion systems that shall enable a stand-aloneinterplanetary CubeSat to escape Earth while min-imising gravity losses, Van Allen belt crossings, andirrecoverable destabilisation.

A mission analysis was performed to calculate thenecessary ∆V for escape while considering finite burnmanoeuvres. The required thrust and burn durationcombinations were calculated to achieve Earth escapein the shortest time while adhering to system con-straints.

Monopropellant thruster design using FLP-106propellant was pursued and the system was sized us-ing the ∆V requirements. The propellant propertieswere analysed, the operating pressures were set andthe corresponding feed system design characteristicswere obtained. The thruster design and performance

IAC–18–C4.6.2.x47764 Page 13 of 15

69th International Astronautical Congress, Bremen, Germany.

calculations were done and a total maximum thrustof 3.18 N (1.54 N each) and an Isp of 259.7 secondswere obtained. The overall system mass was esti-mated to be 5.6 kg while occupying 7U space in the16U spacecraft.

A similar design for the bipropellant thruster usingEthanol and H2O2 was pursued and the correspond-ing propellant masses with an O/F ratio of 4.5 wasobtained. Catalytic decomposition of H2O2 is used togenerate the vapours that will readily ignite with thefuel. The performance analysis yielded a maximumthrust of 2.91 N and a Isp of 303.15 s. The systemweighs 4.85 kg while occupying 6.5U space.

Both systems require precise control methodolo-gies to tune the thrust and burn duration so as toachieve escape within the shortest time. The systemssatisfied the system mass requirement (< 25%) as themonopropellant system weighs 18.7% and the bipro-pellant system weighs 16.17% of the overall space-craft mass (30 kg). The performances and the designcharacteristics of the propulsion systems render theEarth escape and the Mars circularisation using astand-alone interplanetary CubeSat feasible.

References

[1] T. H. Zurbuchen, R. von Steiger, S. Bartalev,X. Dong, M. Falanga, R. Fléron, A. Gregorio, T. S.Horbury, D. Klumpar, M. Küppers, et al., “Perform-ing high-quality science on cubesats,” Space ResearchToday, vol. 196, pp. 11–30, 2016.

[2] F. Topputo, M. Massari, J. Biggs, P. Di Lizia,D. Dei Tos, K. Mani, S. Ceccherini, V. Franzese,A. Cervone, P. Sundaramoorthy, et al., “LUMIO:a cubesat at Earth-Moon L2,” in 4S Symposium,pp. 1–15, 2018.

[3] K. Lemmer, “Propulsion for cubesats,” Acta Astro-nautica, vol. 134, pp. 231–243, 2017.

[4] G. Manzoni, “Design of a highly integrated mi-cropropulsion system for microsatellites attitudecontrol,” in 36th AIAA/ASME/SAE/ASEE JointPropulsion Conference and Exhibit, p. 3476, 2000.

[5] S. S. Board, N. R. Council, et al., Assessmentof Mars science and mission priorities. NationalAcademies Press, 2003.

[6] S. Asmar and S. Matousek, “Mars Cube One(MarCO): the first planetary cubesat mission,” inProceedings of the Mars CubeSat/NanoSat Work-shop, Pasadena, California, November, vol. 20, p. 21,2014.

[7] W. K.-F. Tai, “Mars communication network designtrade study,” Master’s thesis, San Jose State Univer-sity, 1998.

[8] R. Staehle, D. Blaney, H. Hemmati, M. Lo,P. Mouroulis, P. Pingree, T. Wilson, J. Puig-Suari,A. Williams, B. Betts, et al., “Interplanetary cube-sats: opening the solar system to a broad commu-nity at lower cost,” Journal of small satellites, vol. 2,no. 1, pp. 161–186, 2013.

[9] K. Mani, F. Topputo, and A. Cervone, “Dualchemical-electric propulsion systems design for inter-planetary cubesats,” in 6th Space Propulsion Confer-ence, pp. 1–12, 2018.

[10] A. Klesh and J. Krajewski, “MarCO: CubeSats toMars in 2016,” in Proceedings of the 29th AnnualAIAA/USU Small Satellite Conference: SSC15-III-3, 2015.

[11] J. Cardin, K. Coste, D. Williamson, and P. Gloyer,“A cold gas micro-propulsion system for cubesats,”in 17th Annual AIAA/USU Conference on SmallSatellites, 2003.

[12] R. Walker, D. Kochny, C. Bramanti, and I. Carnelli,“Miniaturised Asteroid Remote Geophysical Ob-server (M-ARGO): a stand-alone deep space cube-sat system for low-cost science and exploration mis-sions,” in 6th Interplanetary CubeSat Workshop -Cambridge, iCubeSat, 2017.

[13] C. Scharlemann, M. Schiebl, K. Marhold, M. Taj-mar, P. Miotti, C. Kappenstein, Y. Batonneau,R. Brahmi, and C. Hunter, “Development and testof a miniature hydrogen peroxide monopropellantthruster,” in 42nd AIAA/ASME/SAE/ASEE JointPropulsion Conference & Exhibit, p. 4550, 2006.

[14] D. Scharfe and A. Ketsdever, “A review of highthrust, high delta-v options for microsatellite mis-sions,” in 45th AIAA/ASME/SAE/ASEE JointPropulsion Conference & Exhibit, p. 4824, 2009.

[15] C. Scharlemann and M. Tajmar, “Developmentof propulsion means for microsatellites,” in 43rdAIAA/ASME/SAE/ASEE Joint Propulsion Confer-ence & Exhibit, p. 5184, 2007.

[16] H. Barber, C. Buell, G. Falkenstein, and R. Gurnitz,“Microthrusters employing catalytically reacted N2-O2-H2 gas mixtures, Tridyne,” Journal of Spacecraftand Rockets, vol. 8, no. 2, pp. 111–116, 1971.

[17] J. Mueller, R. Hofer, and J. Ziemer, “Survey ofpropulsion technologies applicable to cubesats,” inJoint Army-Navy-NASA-Air Force (JANNAF), Col-orado Springs, CO; United States, 2010.

[18] M. M. Micci, Micropropulsion for small spacecraft,vol. 187. Aiaa, 2000.

[19] B. Reed, W. d, and L. Dang, “Experimental evalua-tion of cold flow micronozzles,” in 37th Joint Propul-sion Conference and Exhibit, p. 3521, 2001.

[20] F. Maggi, G. Gariani, L. Galfetti, and L. T. DeLuca,“Theoretical analysis of hydrides in solid and hybrid

IAC–18–C4.6.2.x47764 Page 14 of 15

69th International Astronautical Congress, Bremen, Germany.

rocket propulsion,” international journal of hydrogenenergy, vol. 37, no. 2, pp. 1760–1769, 2012.

[21] A. Ketsdever and J. Mueller, “Systems considera-tions and design options for microspacecraft propul-sion systems,” in 35th Joint Propulsion Conferenceand Exhibit, p. 2723, 1999.

[22] F. Topputo and E. Belbruno, “Earth–mars trans-fers with ballistic capture,” Celestial Mechanics andDynamical Astronomy, vol. 121, no. 4, pp. 329–346,2015.

[23] F. Soler Lanagrán, “Low-thrust heliocentric transferwith ballistic capture and orbit circularization for astandalone mars cubesat,” Master’s thesis, Politec-nico di Milano, 2018.

[24] N. Wingborg and J. de Flon, “Characterization ofthe ADN-based liquid monopropellant FLP-106,”Space Propulsion 2010, pp. 3–6, 2010.

[25] A. S. Gohardani, J. Stanojev, A. Demairé, K. Anflo,M. Persson, N. Wingborg, and C. Nilsson, “Greenspace propulsion: Opportunities and prospects,”Progress in Aerospace Sciences, vol. 71, pp. 128–149,2014.

[26] K. Anflo and B. Crowe, “In-space demonstra-tion of an ADN-based propulsion system,” in 47thAIAA/ASME/SAE/ASEE Joint Propulsion Confer-ence & Exhibit, p. 5832, 2011.

[27] M. Wurdak, F. Strauss, L. Werling, H. K. Ciezki,D. Greuel, R. Lechler, N. Wingborg, D. Hasan, andC. Scharlemann, “Determination of fluid propertiesof the green propellant FLP-106 and related materialand component testing with regard to applicationsin space missions,” in Space Propulsion Conference,Bordeaux, France, 2012.

[28] A. Larsson and N. Wingborg, “Green propellantsbased on ammonium dinitramide (ADN),” in Ad-vances in Spacecraft Technologies, InTech, 2011.

[29] M. Wilhelm, M. Negri, H. Ciezki, andS. Schlechtriem, “Preliminary tests on thermalignition of ADN-based liquid monopropellants,”Acta Astronautica, 2018.

[30] N. Wingborg, A. Larsson, M. Elfsberg, andP. Appelgren, “Characterization and ignitionof ADN-based liquid monopropellants,” in 41stAIAA/ASME/SAE/ASEE Joint Propulsion Confer-ence & Exhibit, p. 4468, 2005.

[31] K. Anflo and R. Möllerberg, “Flight demonstrationof new thruster and green propellant technology onthe prisma satellite,” Acta Astronautica, vol. 65,no. 9-10, pp. 1238–1249, 2009.

[32] C. R. Dandekar, Y. C. Shin, and J. Barnes, “Machin-ability improvement of titanium alloy (ti–6al–4v) vialam and hybrid machining,” International Journalof Machine Tools and Manufacture, vol. 50, no. 2,pp. 174–182, 2010.

[33] A. Polishetty, M. Goldberg, G. Littlefair, M. Put-taraju, P. Patil, and A. Kalra, “A preliminary as-sessment of machinability of titanium alloy ti 6al 4 vduring thin wall machining using trochoidal milling,”Procedia Engineering, vol. 97, pp. 357–364, 2014.

[34] J. R. Wertz, D. F. Everett, and J. J. Puschell, “Spacemission engineering: the new smad,” ch. 18, pp. 545–546, Microcosm Press, 2011.

[35] M. Negri, M. Wilhelm, C. Hendrich, N. Wingborg,L. Gediminas, L. Adelöw, C. Maleix, P. Chaber-naud, R. Brahmi, R. Beauchet, et al., “New technolo-gies for ammonium dinitramide based monopropel-lant thrusters–The project RHEFORM,” Acta As-tronautica, vol. 143, pp. 105–117, 2018.

[36] G. P. Sutton and O. Biblarz, Rocket propulsion ele-ments. John Wiley & Sons, 2010.

[37] A. Woschnak, D. Krejci, M. Schiebl, and C. Scharle-mann, “Development of a green bipropellant hy-drogen peroxide thruster for attitude control onsatellites,” Progress in Propulsion Physics, vol. 4,pp. 689–706, 2013.

[38] G. Rarata and P. Surmacz, “The safe preparation ofHTP and concentrated H2O2 samples,” Prace Insty-tutu Lotnictwa, pp. 121–130, 2011.

[39] W. Wieling, B. Zandbergen, A. Mayer, andF. Schretijer, “Development of a hydrogen perox-ide/ethanol thruster for the advanced re-entry ve-hicle,” in 3rd Space Propulsion Conference, 2012.

[40] A. Ponomarenko, “RPA-Tool for Rocket PropulsionAnalysis,” in Space Propulsion Conference, 2014.

[41] B. Zandbergen, “Lecture notes in thermal rocketpropulsion,” Faculty of Aerospace Engineering, DelftUniversity of Technology, 2010.

[42] D. K. Huzel, Modern engineering for design of liquid-propellant rocket engines, vol. 147. AIAA, 1992.

[43] T. E. Tietz and J. W. Wilson, Behavior and prop-erties of refractory metals. Edward Arnold, London,1965.

IAC–18–C4.6.2.x47764 Page 15 of 15